"The social costs of gambling exceed the benefits (including consumer surplus from recreational gambling and tax revenue for governments)."

Overview of poll results by Associate Professor Michael O'Neil

Michael O'Neil

The results were that (N=28) 9 respondents agreed, 9 disagreed and 10 were uncertain.  A number of respondents commented that the question was poorly posed which I think reflects the debate as to what are and how to measure social costs (i.e. private and public costs) of gambling.

Read more

Panelist responses

Panelist responses weighted

See all panelist responses

Name Response Confidence Comment
Peter Abelson

Uncertain (neither agree nor disagree) 7 The proposition is plausible but evidence is needed to support the assertion.
Kym Anderson
Gary Banks

Strongly disagree 8 The best answer is in the Productivity Commission's 2010 report, on which I presided as Chairman. While estimates of the social costs are large, ranging at that time up to $8.4 billion, the social benefits are estimated to be even larger, ranging up to $15.8 billion -- with a minimum of around $12 billion (nearly 50 per cent above the maximum cost estimate).
Of course that does not mean there is no role for policy. The Commission found that the social costs could be greatly reduced, with only a relatively small consequent reduction in social benefits, by adopting certain measures related to poker machine design and pre-commitment technology (the latter an instance of behavioural economics at work). Net benefits would according rise greatly. But of course reducing social costs means reducing losses by 'problem gamblers', who account not only for a large share of industry profits, but also a sizeable proportion of own-state government revenue. So the most effective policy instruments have been ignored or dismissed in favour of such trivial alternatives as making clocks visible, and ATMs less so.
Gary Barrett
Harry Bloch
BLOCH, Harry
Jeff Borland
Disagree 8 I am interpreting agreeing with this statement to mean that the socially optimal level of gambling is zero - and therefore I have disagreed because I don't think that is correct. But there is overwhelming evidence that there is a group (perhaps increasing) of 'problem' gamblers for whom (at their current levels of gambling) the cost to society outweighs the benefit. The major policy issue is therefore to reduce problem gambling while allowing gambling activity to continue. It is a policy issue where there seem to be some clear solutions (that at least could be trialled - see for example recent Productivity Commission reports); but it is another example of an area where most politicians are running scared of lobby groups (sports clubs, licensed clubs and casinos) and also where State governments are clearly concerned about the effect on their revenue. The decision by the previous Abbott government to return decision-making rights over regulating gambling matters such as poker machines to the States was terrible - giving the decision-making rights to the level of government that is most conflicted.
David Butler
Uncertain (neither agree nor disagree) 7 For the great majority, no, the benefits exceed any social costs as with most past times. But for a small minority the costs, private and then social, far exceed the benefits. I don't find the Becker-Murphy 'rational addiction' model a particularly useful guide for public policy. 
Matthew Butlin
BUTLIN, Matthew
Lisa Cameron
Strongly agree 9 Gambling results in a transfer of resources from the most vulnerable in society to the more wealthy, including the mega-wealthy. No amount of tax revenue (and consumer surplus) can compensate for the destruction of some people's lives.
Fabrizio Carmignani
Bruce Chapman
Uncertain (neither agree nor disagree) 10 I don't know what the evidence on this is. I am very confident that I don't know what the evidence is.
Deborah Cobb-Clark
Max Corden
Lin Crase
Kevin Davis
DAVIS, Kevin
Brian Dollery

Uwe Dulleck

Uncertain (neither agree nor disagree) 8 I am not sure whether this statement is well crafted. The basic challenge is how to aggregate welfare across different individuals. Behavioural Economics tells us that gamblers hardest hit from an addiction to play suffer self control problems and most likely a revealed preferences argument does not apply, hence these gamblers do not only cost society but we cannot even be sure that they derive welfare out of this activity. Thus in this case we definitely have social costs that are higher than the benefit. If we assume that cognitive biases do not play a role (not a very reasonable assumption), we end up in a situation where the it all depends on how one aggregates costs and benefits, including government revenue from gambling. This aggregation problem is in my eyes the crux of the problem, who counts (what about family members that are adversely affected), how to compare monetary and non-monetary costs.

To get back to Behavioural Economic arguments, I do feel that "Nudges" in this space can and should play a larger role. Allowing, even forcing, people to set unrestricted limits for the amount of money they want to spend on any night (when they are in cold state of decision making, i.e. before entering a hot state while gambling) are hard to argue against, if one believes in the rationality. Such interventions, would make it much easier to argue that gambling is likely to increase welfare and choices.
Mardi Dungey
Uncertain (neither agree nor disagree) 1 This question is too poorly posed for me to answer it other than on a personal level. I would have to know (or assume) far more about the welfare function that society wanted to maximise and the relative weights put on individual freedom of expression and habit versus minimising social loss due to gambling associated health and welfare problems. Not to mention what we will include in definition of gambling - Keno? slot machines? lotto? share holdings? Where we want to draw the line is a social question also. I think one of the lessons we are learning from these poll exercises is that, as with other sciences, it is important to frame the question relatively precisely as there are a lot of potential variables that can be involved in a very general case. The alternative is that we ask the economists to state their assumptions in making their answers - that would give us an interesting view on the differences in assumptions, frameworks and answers.
Chris Edmond
Henry Ergas
ERGAS, Henry

Saul Eslake

Agree 6  
Allan Fels
FELS, Allan
Agree 8  
Gigi Foster

Disagree 3 Quantifying the social damage for which gambling is directly or indirectly responsible is not easy. As with other "sins" (e.g., drinking, smoking, drug use, prostitution) there are many potential costs and benefits both to individuals and to whole groups, and which of those materialises in any given instance is both hard to measure and dependent on unseen factors like stress levels and cultural norms. Any answer to this question hence requires making a call on the net influence of these background factors, plus a commitment to some taxonomy of good and bad (e.g., is it "good" or "bad" for people to get high once in a while?). As with other sins, the canonical economic advice would be to keep them legal (if regulated) so as to avoid pushing them into a black market, enabling society to see the behaviour: the government can then use that visibility not only to tax the behaviour, but to try to work against the worst of its negative effects.
John Freebairn
Uncertain (neither agree nor disagree) 3 For most recreational gamblers and about a half of dollars spent, gambling is one of many recreation choice options with no problem gambling costs; the other half spent by problem gamblers involves spill-over costs on third parties and personal costs inconsistent with rational choice (Productivity Commission, 2010). To selectively tax or regulate against recreational gamblers, but not other forms of recreation such as restaurants and concerts, distorts these decisions.These distortions have to be balanced against the external costs and individual poor choice decisions by problem gamblers. Numbers to make the comparison are uncertain.

Selective taxation of gambling primarily involves a transfer from gambler to government. Productivity Commission (2010) analysis shows this to be a highly regressive tax. Also, the tax fails horizontal equity with those choosing gambling over alternative recreation and other pursuits on similar incomes/expenditures pay different tax burdens.
Paul Fritjers
Agree 8  

Lata Gangadharan


Agree 9 Depends on how one thinks about social costs. If we were to include the moral costs of gambling and the associated implications for corruption and fraud, then the social costs would be certainly higher than the benefits from increased revenue generation.
Ross Garnaut
Robert Gregory
Stephen King
KING, Stephen
Disagree 7 I interpret the question as whether the TOTAL economic welfare from gambling is negative. While this is an empirical question, the likelihood that the total economic welfare from gambling at the level we see in Australia (which is altered by taxes) is negative seems rather unlikely. A more interesting question would be whether at the MARGIN gambling creates positive or negative economic welfare (i.e. should there be a marginal change in gambling in Australia). But that is not what is asked.
Geoff Kingston
KINGSTON, Geoffrey
Uncertain (neither agree nor disagree) 5 I would agree that watching someone play the pokies is unedifying. On the other hand, we need a certain amount of randomness in our lives--why else do we have children?

Michael Knox

KNOX, Michael
Rodney Maddock
Disagree 6 This is an unusual case where the costs are quite visible but the benefits widely dispersed and very difficult to calculate. Even what we include as gambling is problematic. Stock trading and opening a small business are risky activities with money at stake: are they included in gambling? In our liberal democratic society I believe we should generally allow people to do what they want and be very careful about over-riding their individual choices.
Tony Makin
Uncertain (neither agree nor disagree) 7 It is difficult to answer this question definitively without knowing what comprises "social costs". Presumably they include foregone income, borrowing costs, bankruptcies, family and mental health problems etc, some of which would be hard to quantify in dollar terms.
Warwick McKibbin
Doug McTaggart
Disagree 8 While we might not like gambling, or other people gambling, given our own morals, those who do gamble generally don't impose costs on other people and get quite a lot of benefit from the activity.

Where problem gamblers impose external costs, then it creates quite a problem. There does not seem to be any quantity (quota) or price (tax) instrument that would limit the costs. Trying to ban gambling just doesn't work, and imposes significant other costs. However, my guess is that the costs from these individuals is quite low relative to the benefits others enjoy.
Menezes photo Flavio Menezes
James Morley
Uncertain (neither agree nor disagree) 5 It is difficult to measure social costs or consumer surplus. Gambling provides some utility to its participants, but it is also addictive. In saying definitively whether the costs exceed the benefits, I would worry about implying either support for outright prohibition of all forms of gambling or support for a completely unregulated market. I would think many limits should be placed on gambling to reduce its ability to lead to financial ruin (e.g., allow low-stakes gambling like community-hall bingo, but tax or limit participation in higher-stakes contests). For risk-lovers who particularly seek out high-stakes gambles, there are always financial markets available to participate in and these markets even sometimes have positive expected returns, making them a far superior, if less glamorous, option for satisfying their needs. Also, a Pigouvian tax that helps cover some of the costs placed on society could make sense. But gambling will occur regardless of whether it is in a legal market. So outright prohibition really isn't the solution.
Margaret Nowak
NOWAK, Margaret
Agree 5  
Adrian Pagan
PAGAN, Adrian
John Piggot
Uncertain (neither agree nor disagree) 5  
John Quiggin
Strongly agree 10 For most forms of gambling (lotteries are the main exception), more than 50 per cent of revenue is derived from a small proportion of the population, consisting of current and potential problem gamblers. The damage suffered by this group and their families, employers and others exceeds the benefits generated by the share of revenue (less than 50 per cent) derived from recreational gambling.
Rana Roy
ROY, Rana
Disagree 5 Although the Mahabharata has been a life-long companion to me, I am not familiar with more recent literature on the social costs of gambling. Hence, I cannot answer this question with a high degree of confidence. Nonetheless, I am prepared to “take a punt” and answer it in the negative for the simple reason that the benefits from gambling are vastly under-stated in public discussion. They exceed by far the consumer surplus and tax revenues identified in the question itself. They include inter alia: a rare opportunity for social mobility in a society with high levels of generational inequality and social immobility (cf. Australia’s location in the Great Gatsby Curve as recorded in the paper by Mendolia and Siminski in the September issue of Economic Record); the educational value of witnessing wealth acquired transparently by luck in a society where many are misled to believe that the relative life-chances of plutocrats and paupers are determined by “merit”; the quantifiable benefits of both philanthropic and far-sighted for-profit investments by those who have made their fortunes in gambling and are therefore less likely to let their minds be dulled by conventional thinking (cf. the recent and prospective transformation of Hobart’s landscape resulting from David Walsh’s investments in Mona and other ventures and his proposed investments at Macquarie Point); and the hard-to-quantify but historically demonstrable benefits of permitting religious liberty for all, given that bans on gambling entail a violation of the religious liberty of some (cf. Stanley Wolpert’s discussion of the meaning of the Emperor Aurangzeb’s “war on gambling”). Let me end with a personal story. In 1983, as a postgraduate student with nothing but my debts to depend upon, I sought to borrow £1,000 from my bank to place on a bet at odds of 100 to 1 (a bet on India winning the Cricket World Cup). My bank manager refused – thereby depriving me of £100,000 (≈ $250,000 in 1983 dollars) and, with it, a passport out of poverty. Today, the ease with which I was able to make a little money through internet gambling on the result of the US Presidential election is something for which I am truly grateful.
Peter Sheehan
Jeffrey Sheen
SHEEN, Jeffrey
Disagree 6  
Hugh Sibly
Strongly disagree 8 The question as posed appears  ambiguous  to me. I have interpreted it as saying that the social cost of gambling in Australia necessarily outweighs the benefit. The implication being then that gambling should be prohibited. Currently in Australia there is certainly a significant social cost arising from problem gambling. For instance, there is a great deal of concern remains about the social cost of the over use of pokies and the inappropriate promotion of gambling on broadcast sports events. Such problems certainly lower the social benefit from gambling, and are best controlled through regulation as indicated in the Productivity Commission's 2010 report.
stepledon photo (2)Nigel Stapledon
John Tisdell
Joaquin Vespignani
Agree 7

View our related polls